Re: The End-To-End Internet (was Re: Blocking MX query)

[Eliot Lear <lear () cisco com>]

On 9/6/12 8:27 PM, Owen DeLong wrote:
> Despite my scepticism of the overall project, I find the above claim a
> little hard to accept.  RFC 2052, which defined SRV in an
> experiment, came out in 1996.  SRV was moved to the standards track in
> 2000.  I've never heard an argument why it won't work, and we know
> that SRV records are sometimes in use.  Why couldn't that mechanism be
> used more widely? 
>
> If browsers started implementing it, it could.

This is currently being discussed in the httpbis working group as part
of the http 2.0 discussion.  Also, I'll note that at least one browser
has implemented XMPP without the mandatory SRV record, and it's next to
useless for XMPP (in fact it seems to only work with a handful of broken
XMPP implementations), so look for SRV in at least one browser in the
next year or so, I'd guess.


> I suppose the more accurate/detailed statement would be:
>
> Without modifying every client on the internet, there is no way for the
> clients trying to reach you to reliably be informed of this port number
> situation.
>
> If you're going to touch every client, it's easier to just do IPv6.

Well, this depends on who you think "you" is.  The browser gang
regularly touches many MANY (but not all) clients.

Eliot